OK, this is real basic but I'm not getting it. I hooked up my HP
8657A signal gerator into my TDS 210 oscope with RG 58a/u ,which is 50 ohm to match my input and output. The trouble is , the voltage readings on my scope do not match my output readings on the signal generator., sometimes off by a factor of 3 to 4. The scope is set for
1X probe . Why the hugh mismatch? thanks jk
Since the TDS210 scope does not have 50 ohm input coupling (As many scopes do) the voltage will be double what is expected. As Phil said, get a 50 ohm terminator to properly view the output. If you're not too concerned with signal integrity, you can solder a 50 ohm resistor across a coax T connector.
For a generator described by Vg and Rg (with Rg=Zo as in your case), the transfer function VL/Vg is H=ZL/(Zo+ZL)exp(-theta). In the lossless case (for short cable runs) theta is j*w*l/Vp, with l the cable length [m] and Vp=2E8 [m/s]. In your case ZL=inf and abs(H)=1.
But, the generator usually displays the rms voltage to a matched load. This means that you get twice the voltage for on open circuit (as in your case) and the peak to peak voltage is sqrt(2) times the rms voltage. This means an amplitude 2*1.41 times the displayed value.
If the generator is a 50 ohm source, the end of the coax looks like a
50 ohm source. Hook that to a 15 pF scope input and you get a 1st order lowpass response with corner around 200 MHz or so. No standing waves or resonances at all.
--
For an arbitrary frequency and an arbitrary length of cable, that's
only true if the source impedance, the transmission line impedance,
and the load impedance are all equal and resistive, with the
transmission line looking like an attenuator.
No. Given a 50 ohm generator driving a 50 ohm cable, open on the far end, the end of the coax looks like a 50 ohm source. The transmission line doesn't look like an attenuator unless it's lossy, which effect will be negligable for a few feet of RG58 at reasonable frequencies.
Any reflections will be absorbed by the generator impedance. If you sweep the gen frequency, the scope will see a voltage that looks just like a 1st order RC lowpass response, flat to about 200 MHz and then rolling off at -6 dB per octave. If there are any standing waves, they are somewhere inside the cable, totally invisible from outside; they won't cause frequency peaks or dips at the scope.
LT Spice has a tranny model, so you can try it.
A good signal generator should be a 50 ohm source at all frequencies.
Crazy. I'm considering the scope input to look like a 15 pF cap. If it's different, the 3 dB rolloff point just moves.
Not at all; it will be flat up the the 3dB rolloff point, about 200 MHz in my example. Look up "source termination."
Try this: connect a fast 50 ohm pulse generator, through a length of RG58, to a hi-z scope. A square pulse in will make a square pulse on the scope, no ringing or anything. For long cables with a fast generator/scope, you may see a little rising-edge slowdown from cable losses.
Right. The end of the coax looks like a 50 ohm source, so a 50 ohm load will form a simple voltage divider with gain of 0.5.
The frequency response is the same either way, flat until the scope capacitance or the scope bandwidth starts to roll it off.
RF signal generators are usually spec'd/calibrated to deliver some voltage or power into a 50 ohm load. Function generators are usually spec'd/calibrated to deliver some voltage into an open circuit. The difference is 2:1 voltage.
As Phil said, get a 50
The waveform will be the same, just different amplitude.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.